library_name: pytorch
tags:
- robotics
- world-model
- visual-world-model
- model-based-control
- surface-vehicle
- hidden-drift
FlowMo: Flow-Momentum World Model
FlowMo is a clean-image world-model benchmark for surface vehicles under hidden water drift. The proposed model separates short-history endogenous state and momentum from long-history exogenous drift context, then evaluates whether that factorization improves rollout prediction and closed-loop planning.
This repository currently contains the public code, tests, configuration, and canonical paper datasets. Official checkpoints, generated GIFs, tables, and full experiment reports will be uploaded after the paper-scale training and evaluation runs finish.
Paper Pipeline
Run the complete paper-facing experiment:
python -m experiments.run_paper_image_pipeline
The default command trains all learned world models, evaluates prediction, runs FlowMo latent probes, evaluates planning on all configured tasks and boat morphologies, generates GIFs, and writes:
experiments/reports/paper_prediction_seen_flow_diagnostic.json
experiments/reports/paper_prediction_unseen_flow.json
experiments/reports/paper_prediction_unseen_boat_params.json
experiments/reports/paper_flowmo_latent_probes.json
experiments/reports/paper_planning/
experiments/reports/paper_report.md
Images are rendered online from simulator states. Model inputs are clean top-down RGB frames with no flow arrows, no goal markers, no velocity vectors, and no trajectory overlays.
Compared Methods
flowmo: proposed Flow-Momentum World Model.leworldmodel: LeWorldModel-style JEPA latent predictor.planet: PlaNet-style RSSM world model.tdmpc2: TD-MPC2-style latent dynamics world model.pid_los_controller,physics_mpc_no_flow,current_estimator_mpc,oracle_flow_mpc: traditional planning/control baselines.
Baseline fidelity and naming rules are documented in experiments/BASELINES.md.
The complete paper experiment matrix is documented in experiments/EXPERIMENT_MATRIX.md.
Tests
python -m pytest -q